Expand description

nncombinator is a neural network library that allows type-safe implementation.

Modules

Activation Function Implementation
Array-related data structures such as fixed-length arrays
Function to wrap and handle cuda kernel
Computational processes used in the implementation of neural networks
Definition of various errors
The various layers that make up a neural network and the traits they implement
Implementation of a unidirectional list to be used for memory management in a memory pool
Implementing the loss function of a neural network
Functions related to cuda memory
Definition and implementation of various operations on Units
Definition and implementation of optimizers to be used during training
Implementation on persistence of neural network models

Structs

Stack containing elements
Empty stack, containing no elements

Traits

Trait that defines a stack to store the results computed by forward propagation when training a neural network.